An object tracking method using adaptive template size update in IR image sequence

Abstract

As navigation and surveillance systems are widely introduced, an object tracking method becomes an important component of the systems. The template matching methods have been developed and frequently used for object tracking, which is fast enough, but not robust to a shape- and size-changed object. In order to overcome the limitation of the template matching method, this thesis proposed a template update technique. Firstly, after finding an object position using the correlation-based adaptive predictive search, the proposed method selects blocks which include object’s boundary and have high edge energy. Then, this method judges motion of boundary using block matching. This method finally updates the template considering the boundary block matching. For verifying the proposed method, we applied it to IR image sequences including an approaching object. From the experimental results, the proposed method showed successful performance to track object until final frame.